On 9 April 2017 at 14:33, R A Brown <[log in to unmask]> wrote:

> On 09/04/2017 12:18, And Rosta wrote:
>> On 9 Apr 2017 08:24, "R A Brown" wrote:
>>> The regular vocative singular was _Dee_ which was used by
>>> Tertullian and Prudentius. But pronouncing this as two
>>> syllables (both Es being short) correctly in normal speech
>>> is clearly awkward.
>>> It was IMO fairly 'natural' for the anonymous
>>> translators of the Old latin versions to use _deus_ rather
>>> than the phonetically awkward _dee_.
>> I am away from my copy of Vox Latina, so let me ask more about this
>> awkwardness. Any awkwardness surely can't have been literally purely
>> phonetic.
> It's awkward in English, isn't it?  I can't think of any words where /ɛ.ɛ/
> occur in contiguous syllables.

That's because /ɛ/, the DRESS vowel, is checked -- must be followed by a
consonantal coda. In English, /ɛ.ɛ/ is impossible on phonological grounds.
[ɛ:] is not phonetically difficult (e.g. it happens to be how most English
people of my generation and younger pronounce _air_). Despite the existence
of an IPA syllabicity marker, it would take a lot to persuade me that
syllabicity is a phonetic rather than phonological property.

> But if they do they surely will not retain separate pronunciations in
> allegro speech.


> Assuming that contiguous heterosyllabic (what's the evidence for
>> the heterosyllabicity?) identical nonhigh vowels were phonologically
>> licit,
>> perhaps the problem was the combination of their rarity and the
>> marginality
>> of their phonetic contrast with other long monophthongs?
> Fairly marginal and clearly contracted in normal speech to a single vowel,
> cf. cohors ~ cors.  It's clear from the Romance languages that the
> contracted short vowel was the norm in speech, e.g. cover [verb] <-- Old
> French _covrir_ (Modern _couvrir_) <-- Vulgar Latin *coperīre ( = Classical
> cooperīre).
> Or maybe they were
>> phonologically licit in earlier Latin but became illicit along with other
>> reorganizations to the vowel system? --
> They were marginal and simplified in spoken Latin. Turtullian and
> Prudentius may have used the "correct" _Dee_ in writing.  But had this form
> caught on it would surely have been pronounced just as /dɛ/.  The evidence
> is clear that Latin speaking Jews and, later, Latin speaking Christians
> felt happier with _Deus_ /'dɛʊs/
So let's ask why it became /'dɛʊs/. The phonetic difficulty would have been
the difficulty of having a monosyllabic--disyllabic contrast in a phonetic
monophthong. So given that the phonologically expected vocative would have
been /dɛ/ rather than /dɛ.ɛ/, the vocative would have ended up as a bare
stem. How many vocatives consist of just a bare stem, and how many
vocatives share the form of the nominative? Given that an /ɛ/ suffix was no
longer a phonologically licit option, what was the morphologically least
irregular vocative?